Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
This forum is in permanent archive mode. Our new active community can be found here.
AI & Ultra Hal Assistant.
I'm very much into playing with AI even though the state of the art is, well, lacking. Ultra Hal Assistant is supposedly an award winning piece of software. I have been playing with it and blogged about it today. It's pretty sad. Do you think we'll ever have software that allows you to interact with your computer in a way that's more than basic voice commands or mouse clicking? How many years away is this? Is it another flying car?
Comments
I'd give it about 10-15 years at the current rate of advancement.
On one path you have people who are trying to program systems that are not actually intelligent the way humans are, but have complex enough algorithms that they mimic intelligence. I may not be able to make a human in a computer, but I can write a program that can appear play Pac-Man like a human. This path is interesting because it allows us to get more instant gratification. You end up with a system that appears to be somewhat intelligent, even if it is not. The problem is that I do not think that this path of inquiry will ultimately result in a truly intelligent system.
The other path is the kind of thing people are doing working on neurological simulations and quantum computing. I think this path of inquiry does have a much better chance of eventually producing an intelligent system, even though it's a long way off. However, it is very hard to tell if you are making progress since you won't be able to produce any tangible results until you figure it out. It can be much harder to get funding for something like this as you are less likely to produce anything tangible and recognizably useful unless you succeed.
One path goes around the mountain collecting goodies, but never reaches the top. The other path goes straight to the top, is very difficult to climb, but at the end of the path is a great treasure.
I personally believe that true artificial intelligence is far out of our reach and might not be accomplished ever. However, with the constant growth of computing power I think the end of the first path Scott described could be reached within the next 200 years.